A post argues that much of AI labs’ alarming messaging was aimed at attracting and retaining machine-learning researchers by signaling belief in artificial general intelligence and a commitment to handling it responsibly. It credits Anthropic’s positioning on cutting-edge AI and safety with attracting top talent, including from OpenAI, and enabling its technical catch-up with OpenAI. The author’s argument: those signals had to be public and costly to be credible, but labs now have less control over how politicians and the public interpret them. The author stresses that this is about changing incentives—not a claim that AI risks were exaggerated.